Dataset: Large-scale Urban IoT Activity Data for DDoS Attack

Abstract

As IoT deployments grow in scale for applications such assmart cities, they face increasing cyber-security threats. Inparticular, as evidenced by the famous Mirai incident andother ongoing threats, large-scale IoT device networks areparticularly susceptible to being hijacked and used as bot-nets to launch distributed denial of service (DDoS) attacks.Real large-scale datasets are needed to train and evaluatethe use of machine learning algorithms such as deep neu-ral networks to detect and defend against such DDoS attacks.We present a dataset from an urban IoT deployment of 4060nodes describing their spatio-temporal activity under benignconditions. We also provide a synthetic DDoS attack genera-tor that injects attack activity into the dataset based on tun-able parameters such as number of nodes attacked and dura-tion of attack. We discuss some of the features of the dataset.We also demonstrate the utility of the dataset as well as oursynthetic DDoS attack generator by using them for the train-ing and evaluation of a simple multi-label feed-forward neu-ral network that aims to identify which nodes are under at-tack and when.